def concatenate_input_bams(config, input_bams, out = AtomicCmd.PIPE):
"""Transparent concatenation of input BAMs.
Return a tuple containing a list of nodes (0 or 1), and an
object which may be passed to the IN_STDIN of an AtomicCmd
(either an AtomicCmd, or a filename). This allows transparent
concatenation when multiple files are specified, while
avoiding needless overhead when there is only 1 input file."""
input_bams = safe_coerce_to_tuple(input_bams)
if len(input_bams) == 1:
return [], input_bams[0]
jar_file = os.path.join(config.jar_root, "MergeSamFiles.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
params.set_kwargs(CHECK_JAR = _picard_version(jar_file))
if out == AtomicCmd.PIPE:
params.set_kwargs(OUT_STDOUT = out)
params.set_option("OUTPUT", "/dev/stdout", sep = "=")
else:
params.set_option("OUTPUT", out, sep = "=")
params.set_option("CREATE_INDEX", "False", sep = "=")
params.set_option("COMPRESSION_LEVEL", 0, sep = "=")
for (index, filename) in enumerate(safe_coerce_to_tuple(input_bams), start = 1):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep = "=")
params.set_kwargs(**{"IN_BAM_%02i" % index : filename})
params.set_option("SO", "coordinate", sep = "=", fixed = False)
cmd = params.finalize()
return [cmd], cmd
def customize(cls, reference, infiles, outfile, options, dependencies = ()):
infiles = safe_coerce_to_tuple(infiles)
jar_file = os.path.join(options.jar_root,"GenomeAnalysisTK.jar")
UnifiedGenotyper = AtomicJavaCmdBuilder(options,jar_file)
UnifiedGenotyper.set_option("-R", "%(IN_REFERENCE)s")
UnifiedGenotyper.set_option("-T", "UnifiedGenotyper")
for bam in infiles:
assert os.path.exists(bam), "Couldn't find input BAM: {}".format(bam)
UnifiedGenotyper.add_option("-I", bam)
UnifiedGenotyper.set_option("-o", "%(OUT_VCFFILES)s")
UnifiedGenotyper.set_option("-stand_call_conf", "30.0")
UnifiedGenotyper.set_option("-stand_emit_conf", "10.0")
UnifiedGenotyper.set_option("-dcov", "200")
#UnifiedGenotyper.set_option("-nct", "3")
UnifiedGenotyper.set_option("-L", "chrUn2:1-19213991")
UnifiedGenotyper.set_kwargs(
IN_REFERENCE = reference,
OUT_VCFFILES = outfile,
OUT_VCF_IDX = outfile + ".idx"
)
return {
"commands" : {
"unifiedgenotyper" : UnifiedGenotyper
}
}
def customize(cls, config, input_bams, output_bam, output_metrics = None, dependencies = ()):
jar_file = os.path.join(config.jar_root, "MarkDuplicates.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
# Create .bai index, since it is required by a lot of other programs
params.set_option("CREATE_INDEX", "True", sep = "=")
params.set_option("OUTPUT", "%(OUT_BAM)s", sep = "=")
params.set_option("METRICS_FILE", "%(OUT_METRICS)s", sep = "=")
input_bams = safe_coerce_to_tuple(input_bams)
for (index, filename) in enumerate(input_bams):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep = "=")
params.set_kwargs(**{("IN_BAM_%02i" % index) : filename})
# Remove duplicates from output by default to save disk-space
params.set_option("REMOVE_DUPLICATES", "True", sep = "=", fixed = False)
params.set_kwargs(OUT_BAM = output_bam,
OUT_BAI = swap_ext(output_bam, ".bai"),
OUT_METRICS = output_metrics or swap_ext(output_bam, ".metrics"),
CHECK_JAR = _picard_version(jar_file))
return {"command" : params,
"dependencies" : dependencies}
def customize(cls,
config,
input_bams,
output_bam,
output_metrics=None,
dependencies=()):
jar_file = os.path.join(config.jar_root, "MarkDuplicates.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
# Create .bai index, since it is required by a lot of other programs
params.set_option("CREATE_INDEX", "True", sep="=")
params.set_option("OUTPUT", "%(OUT_BAM)s", sep="=")
params.set_option("METRICS_FILE", "%(OUT_METRICS)s", sep="=")
input_bams = safe_coerce_to_tuple(input_bams)
for (index, filename) in enumerate(input_bams):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep="=")
params.set_kwargs(**{("IN_BAM_%02i" % index): filename})
# Remove duplicates from output by default to save disk-space
params.set_option("REMOVE_DUPLICATES", "True", sep="=", fixed=False)
params.set_kwargs(OUT_BAM=output_bam,
OUT_BAI=swap_ext(output_bam, ".bai"),
OUT_METRICS=output_metrics
or swap_ext(output_bam, ".metrics"),
CHECK_JAR=_picard_version(jar_file))
return {"command": params, "dependencies": dependencies}
def concatenate_input_bams(config, input_bams, out=AtomicCmd.PIPE):
"""Transparent concatenation of input BAMs.
Return a tuple containing a list of nodes (0 or 1), and an
object which may be passed to the IN_STDIN of an AtomicCmd
(either an AtomicCmd, or a filename). This allows transparent
concatenation when multiple files are specified, while
avoiding needless overhead when there is only 1 input file."""
input_bams = safe_coerce_to_tuple(input_bams)
if len(input_bams) == 1:
return [], input_bams[0]
jar_file = os.path.join(config.jar_root, "MergeSamFiles.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
params.set_kwargs(CHECK_JAR=_picard_version(jar_file))
if out == AtomicCmd.PIPE:
params.set_kwargs(OUT_STDOUT=out)
params.set_option("OUTPUT", "/dev/stdout", sep="=")
else:
params.set_option("OUTPUT", out, sep="=")
params.set_option("CREATE_INDEX", "False", sep="=")
params.set_option("COMPRESSION_LEVEL", 0, sep="=")
for (index, filename) in enumerate(safe_coerce_to_tuple(input_bams),
start=1):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep="=")
params.set_kwargs(**{"IN_BAM_%02i" % index: filename})
params.set_option("SO", "coordinate", sep="=", fixed=False)
cmd = params.finalize()
return [cmd], cmd
def customize(cls, config, input_bams, output_bam, dependencies=()):
jar_file = os.path.join(config.jar_root, "MergeSamFiles.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
params.set_option("OUTPUT", "%(OUT_BAM)s", sep="=")
params.set_option("CREATE_INDEX", "True", sep="=")
params.set_kwargs(OUT_BAM=output_bam,
OUT_BAI=swap_ext(output_bam, ".bai"),
CHECK_JAR=_picard_version(jar_file))
for (index, filename) in enumerate(input_bams, start=1):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep="=")
params.set_kwargs(**{("IN_BAM_%02i" % index): filename})
params.set_option("SO", "coordinate", sep="=", fixed=False)
return {"command": params, "dependencies": dependencies}
def customize(cls, config, input_bams, output_bam, dependencies = ()):
jar_file = os.path.join(config.jar_root, "MergeSamFiles.jar")
params = AtomicJavaCmdBuilder(config, jar_file)
params.set_option("OUTPUT", "%(OUT_BAM)s", sep = "=")
params.set_option("CREATE_INDEX", "True", sep = "=")
params.set_kwargs(OUT_BAM = output_bam,
OUT_BAI = swap_ext(output_bam, ".bai"),
CHECK_JAR = _picard_version(jar_file))
for (index, filename) in enumerate(input_bams, start = 1):
params.add_option("I", "%%(IN_BAM_%02i)s" % index, sep = "=")
params.set_kwargs(**{("IN_BAM_%02i" % index) : filename})
params.set_option("SO", "coordinate", sep = "=", fixed = False)
return {"command" : params,
"dependencies" : dependencies}
